c
SAB 2020
The VISCACHA survey
Age and metallicity spatial distribution of star clusters in the SMC reveals a complex
tidal history
B. Dias
1,2, L. Kerber
3, J. F. C. Santos Jr.
4, F. Maia
4,5, E. Bica
6, B. Barbuy
5, C. Parisi
7,8, D. Geisler
91 European Southern Observatory, Alonso de Córdova 3107, Vitacura 19001, Chile
2 Departamento de Física, Facultad de Ciencias Exactas, Universidad Andrés Bello, Av. Fernandez Concha 700, Las Condes, Santiago, Chile
3 Universidade Estadual de Santa Cruz, Rodovia Jorge Amado km 16, Ilhéus 45662-000, Bahia, Brazil 4 Departamento de Física, ICEx - UFMG, Av. Antônio Carlos 6627, 31270-901, Brazil
5 Departamento de Astronomia, IAG - USP, Rua do Matão 1226, 05508-090, Brazil 6 Departamento de Astronomia, IF - UFRGS, Av. Bento Gonçalves 9500, 91501-970 ,Brazil
7 Observatorio Astronómico, Universidad Nacional de Córdoba Laprida 854, Córdoba, CP 5000, Argentina. 8 Consejo Nacional de Investigaciones Científicas y Técnicas Av. Rivadavia 1917, Buenos Aires, CP 1033, Argentina 9 Departamento de Astronomía, Universidad de Concepción Casilla 160-C, Concepción, Chile.
Abstract. During the past few decades, star clusters were used to shed light on the disordered evolution of the Small Magellanic Cloud (SMC). Age and metallicity radial gradients are good tracers for that, but hard to define for the SMC that has been disrupted due to interactions with the Large Magellanic Cloud (LMC) and the Milky Way. We propose to split the SMC into four groups and show that they reveal different chapters of the SMC evolution history. The groups are main body, wing/bridge, counter-bridge, and west halo. In particular, the west halo has a clear positive trend in age and negative in metallicity. The main body and counter-bridge clusters show a similar pattern. The wing/bridge exhibit a V-shape distribution for metallicity and the opposite for age, as also shown by other works. These tendencies would be more disperse and hard to identify if all components were analysed without distinction. Comparison with the predicted stellar distribution from dynamical models show that the west halo could be a tidal structure disrupted after an interaction with the LMC. In fact, recent proper motion maps of the SMC from HST+Gaia DR1 and from the VMC survey confirmed that the west halo is really being stripped out from the main body. Recently published Gaia DR2 has proper motions for the SMC that should be analysed soon. In this talk we present age and metallicity derived via synthetic colour-magnitude diagram fitting of SMC star clusters observed with the SOAR telescope (joint project between Chile and Brazil) and discuss the results above. Our ongoing VISCACHA survey has already used over 200h of SOAR time and will deliver accurate age, metallicity, distance, and reddening for about 100 star clusters in the SMC. We also started a spectroscopic follow-up using GMOS@Gemini to derive spectroscopic metallicities and radial velocities for key clusters.
Resumo. Durante as últimas décadas, aglomerados de estrelas foram usados para esclarecer a evolução desordenada da Pequena Nuvem de Magalhães (SMC, em inglês). Gradientes radiais de idade e metalicidade são bons traçadores para isso, mas difíceis de definir para a SMC que tem sido distorcida devido a interações com a Grande Nuvem de Magalhães (LMC, em inglês) e a Via Láctea. Propomos uma divisão da SMC em quatro grupos que revelam capítulos diferentes da história de evolução da SMC. Os grupos são: corpo principal, asa/ponte, contra-partida da ponte e halo oeste. Em particular, o halo oeste possui uma clara tendência positiva em idade e negativa em metalicidade. Os aglomerados do corpo principal e da contra-partida da ponte apresentam um padrão similar. A asa/ponte mostra uma distribuição em ‘V’ para metalicidade e o oposto para idade, como mostrado por outros trabalhos. Essas tendências seriam mais dispersas e difíceis de identificar se todas as componentes fossem analisadas sem distinção. Uma comparação com a distribuição estelar prevista por modelos dinâmicos revelam que o halo oeste poderia ser uma estrutura de maré removida após uma interação com a LMC. De fato, mapas recentes de movimento próprio da SMC usando HST+Gaia DR1 e o levantamento VMC confirmaram que o halo oeste realmente está sendo destroncado do corpo principal. O Gaia DR2 foi recentemente publicado e possui movimentos próprios para a SMC que deveriam ser analisados em breve. Nesta palestra apresentamos idades e metalicidades determinadas via ajuste de diagrama cor–magnitude de aglomerados estelares da SMC observados com o telescópio SOAR (projeto conjunto entre Chile e Brasil) e discutimos os resultados acima. O levantamento VISCACHA — em andamento — já usou mais 200h de tempo do SOAR e entregará idade, metalicididade, distância e avermelhamento precisos para aproximadamente 100 aglomerados estelares na SMC. Também demos seguimento à determinação de metalicidades e velocidades radiais via espectroscopia usando o GMOS@Gemini.
Palavras-chave. (galaxies:) Magellanic Clouds — galaxies: star clusters: general — galaxies: stellar content — galaxies: structure — galaxies: evolution
1. Introduction
The VISCACHA (VIsible Soar photometry of star Clusters in tApii and Coxi HuguA1) project (Maia et al. 2019) is a deep pho-tometric survey observing star clusters in the Large and Small
1 LMC and SMC names in the Tupi-Guarani language
Magellanic Clouds (LMC, SMC). In contrast with other large area sky surveys pointing at the Magellanic system — such as the VMC (Cioni et al. 2011) or the SMASH (Nidever et al. 2017) — our survey is focused on star clusters throughout the SMC, LMC, and Bridge. Our advantage is to obtaining spatial reso-lution about five times higher than the large surveys (see Fig.
1) and accurate photometry reaching a few magnitudes fainter than the larger surveys in the crowded regions such as star clus-ters (see Sect. 3). We also use a 4m class telescope as most of the modern surveys, but we combine with the use of adaptive optics (AO) that makes the VISCACHA data to be unique and complementary to any other previous photometric survey on the Magellanic Clouds.
SOI @ SOAR
SAMI @ SOAR
Figura 1. Coloured images of NGC 152 based on BV images taken with SOI and VI images taken with SAMI. The FOV of the images is about 3 × 30. The insets highlights the five times better spatial resolution al-lowed by the AO with SAMI enabling the deep photometry even in the dense cores of the most massive clusters.
6 4 2 0 −2 −4 −6 −6 −4 −2 0 2 4 6 2o 4o 6o 8 o10 o This work Paper I Main body West Halo Wing/Bridge Counter−Bridge E N LMC relative α ⋅ cosδ (o) relativ e δ ( o )
Figura 2. Figure extracted from Dias et al. (2016) showing the defini-tion of the four regions on the projected distribudefini-tion of SMC star clus-ters from the catalogue of Bica et al. (2008): main body, wing/bridge, counter-bridge, and west halo.
The SOuthern Astrophysical Research (SOAR) telescope is accessible to our team through Brazil and Chile hosting together 40% of the available nights. We are using Brazilian and Chilean time to accomplish our planned observations using the SOAR Adaptive Module Imager (SAMI) since its commissioning in 2015. Some of the members are from Argentina, that has access to Gemini telescope, and we are using joint Brazilian, Chilean, and Argentinean time for spectroscopic follow-up observations with GMOS.
We discuss here one of the many projects that are being de-veloped within the collaboration. The SMC star cluster popu-lation is less studied than the LMC popupopu-lation possibly because their spatial distribution is very complex. Dias et al. (2014, 2016) claimed to have find a way to classify the star cluster in groups and start solving the puzzle.
2. The SMC star cluster groups
Parisi et al. (2009, 2015) derived accurate and homogeneous metallicities from CaII lines of red giant stars for 36 SMC clus-ters and characterized the radial metallicity distribution as a V-shape with vertex at about a ∼ 4 − 5◦. This peculiar distribution could be identified also by Piatti (2011) and Dias et al. (2014, 2016) in age and metallicity with a large dispersion. Combined with that, all the four works present an age-metallicity relation (AMR) with a large dispersion in metallicity that seems to be real. Dias et al. (2014) defined four regions on the projected dis-tribution of SMC star clusters as shown in Fig. 2 and indicated that the V-shape on the radial gradients could be originated by a superposition of simple gradients of each group, that are clearly identified when analysed in isolation (see Sect. 4). They also show that the metallicity dispersion seen in the AMR is signifi-cantly reduced if the groups are treated individually. These four groups have physical motivations behind them and each group may be telling a different chapter of the SMC tidal evolution history.
The definition of the SMC main body follows the distribu-tion of young stars by Bekki & Chiba (2009), as can be seen in
−5 0 5
−5
0
5
distance from the centre (kpc)
distance from the centre (kpc)
Clusters younger than 2.7Gyr Simulated stars younger than 2.7Gyr
Figura 3. Figure adapted from Bekki & Chiba (2009) showing their re-sults of N-body simulations for the actual distribution of stars younger than 2.7Gyr in the SMC. We overplot the star cluster positions as in Fig. 2 with the same age interval.
Fig. 3. In other words, recent star formation is happening in this central region. Their simulations also show that the older stars should have a spherical distribution with a central concentration (well characterized with the VMC data, see Rubele et al. 2018), but this is not the case of the star cluster distribution (Fig. 4). A possible cause for this difference is cluster dissolution in the dense regions of the SMC, taking into account that most of these clusters are low-mass.
The wing/bridge region is defined by the gas structure that connects the SMC to the LMC where a stellar counterpart is known. In fact, the wing/bridge clusters follow the Magellanic Bridge of gas very well in Fig. 5. The tidal counterpart of the Bridge is diametrically opposed to it along the line of sight, i.e., it is concentrated on the projected gas distribution in the North of the SMC, as predicted by the simulations of Diaz & Bekki (2012). All clusters in this area were classified as counter-bridge clusters by Dias et al. (2014).
The remaining group on the Southwest region of the SMC was not foreseen by any simulations and not characterized as a group by any observations until a few years ago. It was classi-fied and studied for the first time by our group and denominated ‘west halo’ (Dias et al. 2014, 2016). We found that the west halo clusters have a radial gradient in age and metallicity similar to the gradient found in the inner two degrees of the main body. We compared their distribution on sky with the dynamical sim-ulations of Besla (2011) to conclude that the west halo could be a group of clusters tidally removed from the main body. Proper motions were needed to confirm this scenario.
Recently, two independent studies confirmed the findings of Dias et al. (2016) on the west halo. Niederhofer et al. (2018) published a map of proper motions relative to the bulk SMC motion for the inner regions of the SMC based on VMC data (see Fig. 6). They found no trace of rotation as it is the case of the LMC and found that all vectors in the west halo region
−5 0 5
−5
0
5
distance from the centre (kpc)
distance from the centre (kpc)
Clusters older than 2.7Gyr Simulated stars older than 2.7Gyr
Figura 4. Same as Fig. 3 but for stars older than 2.7Gyr,
10 0 −10 −20 10 0 −10 −20 −30 LMS ( o ) BMS (o) LMC Magellanic Stream SMC Counter−bridge Bridge
Figura 5. Figure adapted from Diaz & Bekki (2012) showing their re-sults of N-body simulations for the actual distribution of gas in the Magellanic System, and we identify the components defined in their work. The star cluster positions are overplotted as in Fig. 2 making clear the position of the wing/bridge and counter-bridge clusters.
are pointing outwards, which is the confirmation that west halo clusters are moving away from the main body. A few months later, Zivick et al. (2018) published a complementary proper motion map based on specific fields observed with HST com-bined with Gaia DR1 spread over a larger area (see Fig. 7). They also showed that the Southern stars are moving from the main
Figura 6. Figure adapted from Niederhofer et al. (2018), showing the proper motion vectors relative to the bulk proper motion of the SMC, based on VMC data. We overplot the four groups of star clusters as defined in Fig. 2.
Figura 7. Similar to Fig. 6 but with the proper motion vectors and base image adapted from Zivicky et al. (2018) using data from HST+Gaia DR1.
body towards the west halo. What calls the attention is that the magnitude of the proper motion vectors is of the same order as those from the Magellanic Bridge, that is known to be a tidal fea-ture of the interaction between LMC and SMC. This means that the bulk motion of the west halo is not negligible. Zivick et al. also pointed out that the vectors of the counter-bridge region are smaller than those of the wing/bridge region possibly because the main component of their motion is along the line of sight, as discussed above.
Figura 8. Figure extracted from Maia et al. (2019) showing the ob-served CMD of Kron 37 with the membership probability of each star indicated by the colour code. The corner plot shows the results of the MCMC Bayesian fit of the CMD to a grid of synthetic CMDs built from PARSEC isochrones (Bressan et al. 2012). The best isochrone fitted to the data is shown as the blue line and the respective synthetic CMD as grey points behind the observed CMD.
3. Colour-magnitude diagram (CMD) fitting: age, metallicity, distance, reddening
One of the main goals of the VISCACHA survey is to trace a 3D map of age and metallicity of SMC star clusters and better characterize the radial gradients and AMR for each group dis-cussed above, possibly constraining the SMC tidal history. The survey has the advantage of using homogeneous and deep obser-vations combined with robust method to derive age, metallicity, distance, and reddening for each cluster.
Kerber et al. (2002); Kerber & Santiago (2005) developed a method to generate a grid of synthetic CMDs for a suitable space of parameters to fit HST deep CMDs of LMC clusters. This method was adapted to SOI@SOAR data by Dias et al. (2014, 2016) and further developed to include Markov chain Monte Carlo (MCMC) method to use Bayesian statistics to fit the observed VISCACHA CMDs to the grid of synthetic CMDs (Maia et al. 2019). This method has been proven to produce ro-bust ages for clusters from a few Myr down to the oldest clus-ters of the Magellanic Clouds 10-11 Gyr old; robust metallicities compatible with those obtained via spectroscopy, even though not as precise (e.g. Dias et al. 2016; Maia et al. 2019); and ac-curate distances comparable to those obtained via other meth-ods (de Grijs & Bono 2015). Reddening is very low in this sky area and agrees with classical reddening maps. Our method de-rives the four parameters altogether without fixing any of them. This has been a common practice in other works, which pro-duces larger systematic errors on age when they assume distance and metallicity, for example.
Fig. 8 shows an example of a CMD fitting for the wing/bridge cluster Kron 37 done by Maia et al. (2019). The ob-served CMD reaches four magnitudes below the main sequence turnoff and has a good number of subgiant, red giant branch and red clump stars, making it a perfect case to produce accurate pa-rameters. The colour scale indicates the probability of each star to belong to the cluster, that is calculated statistically comparing them with a CMD of nearby field outside the cluster tidal radius.
The corner plot displays the results of the MCMC simulations for the four parameters where the maximum likelihood peaks can be easily identified as the final results for this cluster.
4. Discussion and perspectives
We made a schematic cartoon plot of the age and metallicity radial distributions of SMC star clusters older than 1 Gyr for each region (see Fig. 9), based on the literature compilation done by Dias et al. (2014). The interaction that created the Magellanic Stream took place about 1 Gyr ago (Besla et al. 2010) and a substantial population of clusters has been formed since then not only on the main body (e.g. Glatt et al. 2010). Therefore we are not considering the very young cluster population because they are tracing a different history.
The west halo, wing/bridge and counter-bridge regions host the intermediate-age and old clusters being good tracers of the past tidal interactions. In fact, these groups may give constraints to the debate between the classical scenario where the LMC and SMC are orbiting the Milky Way with a period of ∼2 Gyr (e.g. Diaz & Bekki 2012) and the new scenario where the Clouds are in their first infall or have an eccentric orbit with a period > 6 Gyr (e.g. Besla et al. 2010). Marking the stellar populations and star formation bursts that happen during a close encounter between the LMC and SMC and also between the Magellanic system and the Milky Way is certainly one way to constrain these scenarios. Accurate and homogeneous ages, metallicities, and distances are required. Past surveys only reached the clus-ter population younger than 1 Gyr, and present large surveys do not reach old clusters in crowded regions. We intend to present a complete set of parameters for the clusters on the outskirts of the SMC and trace its complete history.
Compilations from the vast literature on this field pro-vides a lot of dispersion because of systematic differences be-tween the methods, photometric filters, large uncertainties etc. Nevertheless, we show in Fig. 9 that the western regions con-taining the west halo and counter-bridge present a clear trend (or gradient) similar to that found on the main body, while the east-ern regions of the wing/bridge reveals a V-shape distribution. If all groups are seen together it looks like that the overall cluster population is distributed in a V-shape. Apparently, the four SMC groups seen on the projected distribution on sky really have dif-ferent characteristics and possibly are tracing different episodes of the SMC tidal history. Moreover, Dias et al. (2014, 2016) pointed to a different AMR for each group, which could indicate merger episodes in each region, or a different chemical enrich-ment history for each region. The VISCACHA survey should provide constraints to these and many more open questions that complement the studies of the large surveys in the field. Agradecimentos. Based on observations obtained at the Southern Astrophysical Research (SOAR) telescope, which is a joint project of the Ministério da Ciência, Tecnologia, e Inovação (MCTI) da República Federativa do Brasil, the U.S. National Optical Astronomy Observatory (NOAO), the University of North Carolina at Chapel Hill (UNC), and Michigan State University (MSU). The au-thors thank the Brazilian and Chilean NTAC for accepting our proposals to carry out VISCACHA observations with SOAR.
References
Besla, G. 2011, PhD thesis, Harvard University
Besla, G., Kallivayalil, N., Hernquist, L., et al. 2010, ApJ, 721, L97
Bica, E., Bonatto, C., Dutra, C. M., & Santos, J. F. C. 2008, MNRAS, 389, 678 Cioni, M.-R. L., Clementini, G., Girardi, L., et al. 2011, A&A, 527, A116 de Grijs, R. & Bono, G. 2015, AJ, 149, 179
Dias, B., Kerber, L., Barbuy, B., Bica, E., & Ortolani, S. 2016, A&A, 591, A11 Dias, B., Kerber, L. O., Barbuy, B., et al. 2014, A&A, 561, A106
✵ ✷ ✹ ✻ ✽ ✶ ✵ ✶ ✷ ❛ ✁ ✂ ✄ ☎ ✆ ✝ ✥✞✟✠ ❜✡☛☞ ✌❡✍✎ ❍✞✏✡ ✑✡✒✠✎❡✓ ❇✓✟☛✐❡ ❇✓✟☛✐❡ ✌✟✠✐ ✔ ✕ ✖ ✗ ✘ ✙✔ ✲ ✶ ✚ ✛ ✲ ✶ ✚ ✵ ✲ ✵ ✚ ✛ ✵ ✚ ✵ ✥✞✟✠ ❜✡☛☞ ✌❡✍✎ ❍✞✏✡ ✑✡✒✠✎❡✓ ❇✓✟☛✐❡ ❇✓✟☛✐❡ ✌✟✠✐ ❬ ✜ ✁ ✢ ✣ ✤ ✦ ❙ ▼❈ ✧ ♦ ✮
Figura 9. Schematic plot of age and metallicity radial distribution based on the literature compilation done by Dias et al. (2014) for clusters older than 1 Gyr.
Diaz, J. D. & Bekki, K. 2012, ApJ, 750, 36
Glatt, K., Grebel, E. K., & Koch, A. 2010, A&A, 517, A50 Kerber, L. O. & Santiago, B. X. 2005, A&A, 435, 77
Kerber, L. O., Santiago, B. X., Castro, R., & Valls-Gabaud, D. 2002, A&A, 390, 121
Maia, F. F. S., Kerber, L. O., Santos Jr., J. F. C., Dias, B., & the VISCACHA team. 2019, MNRAS, submitted
Nidever, D. L., Olsen, K., Walker, A. R., et al. 2017, AJ, 154, 199 Niederhofer, F., Cioni, M.-R. L., Rubele, S., et al. 2018, A&A, 613, L8 Parisi, M. C., Geisler, D., Clariá, J. J., et al. 2015, AJ, 149, 154
Parisi, M. C., Grocholski, A. J., Geisler, D., Sarajedini, A., & Clariá, J. J. 2009, AJ, 138, 517
Piatti, A. E. 2011, MNRAS, 418, L69
Rubele, S., Pastorelli, G., Girardi, L., et al. 2018, MNRAS, 478, 5017 Zivick, P., Kallivayalil, N., van der Marel, R. P., et al. 2018, ApJ, 864, 55
